JPH0410581A - Optical position detection semiconductor device - Google Patents

Abstract

PURPOSE:To obtain a two-dimensional PSD excellent in position detection characteristic by a method wherein a field effect transistor is interposed between a second conductivity type layer region which functions as a photocurrent splitting layer and a belt-shaped electrode so as to electrically connect them together or to electrically disconnect them from each other. CONSTITUTION:When a positive voltage is applied onto a back electrode 5 on the basis of current lead-out electrodes 12, 13, 14, and 15 which are kept at the same potential to start a PSD operating, a positive voltage is applied to gate electrodes 18 and 19 to electrically separate the electrodes 12 and 13 from a current splitting layer 8, an optical position detector serves equivalent to a one-dimensional PSD, which is provided with only the lead-out electrodes 14 and 15 and detects a vertical position, and positional coordinates are accurately calculated. When the positional coordinates of a light beam in a direction vertical to it is obtained, a positive voltage is applied to gate electrodes 16 and 17 the same as above to electrically separate the electrodes 14 and 15 from the current splitting layer 8, then the optical position detector serves equivalent to a one-dimensional PSD which detects a horizontal position and positional coordinates are precisely calculated. In result, a two dimensional optical position detector excellent in characteristics can be obtained.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a semiconductor device for optical position detection that detects the position of incident light or radiation m (hereinafter referred to as light) irradiated onto the surface of a semiconductor element. .

[Prior art] Semiconductor optical beam position detection element (hereinafter referred to as PSDPositi)
on 5ensistive Detector)
is a type of semiconductor sensor whose main function is to detect the position of a light beam irradiated onto the surface of a semiconductor element.

The principle of light beam position detection by PSD utilizes lateral photo effect on the semiconductor surface, and its outline is as follows.

In other words, a current is generated in the semiconductor device by the energy of the light incident on the device, and when this current flows toward at least two or more electrodes provided at the end of the device, the magnitude of the current increases. Since it changes in inverse proportion to the distance to the inverted electrode, the incident light beam position can be determined by measuring the magnitude of the generated photocurrent at each electrode separately and performing a simple calculation.

FIG. 6(a) is a plan view for explaining the conventional PSD,
The same figure (b) is a B-BIIr plane view of the same figure (a). In FIG. 6, a square 33 of a P-type layer 32 is formed on the first surface of a low-concentration n-type semiconductor wafer 31, and this square 33 is used as a light-receiving surface. A high concentration p'' region 34.35 is formed along two opposing sides, and a metal electrode 37.38 is formed on the surface of the p1 region 34.35.
is provided as an output electrode, and a transparent insulating layer 36 is formed on the first surface of the square 33 other than this electrode. Then, an n3 layer 39 is formed on the second surface behind the first surface of the n-type semiconductor wafer 1, and an electrode 4 is formed on a part of this n'' layer 39.
It is set to 0.

As a result, the PSD in Figures 6(a) and (b) has the origin 0 at the center of the light-receiving surface, and the X
If the Y-axis is parallel to the electrodes 37.38, and the distance between the electrodes 37 and 38 is 2L, then when a point of light is incident on the point Q (x, y) of the light receiving surface, each electrode 37.38 The flowing current, I8, is IT +r, =I.

Then, I 7 = I o (l x / L) / 2I a
"Io (1+X/L)/2, and Ia I7=Io"x/Lx-L (18
From IT)/IO, the position coordinate X is x=L (Ia IT)/(Ia +I7)"-"
Determined from (+).

Further, as shown in FIG. 6(c), on the two layers 33 of the PSD, there are high concentration P'''' regions facing each other along two sides 41.42 different from the sides of the electrodes 37.38, and the surface thereof. metal electrode 4
It is possible to function as a two-dimensional PSD by providing 3.44 and determining the y-coordinate using the same principle.

[Problem to be solved by the invention] However, in the case of this two-dimensional PSD, the electrode 37
．． 38 and electrodes 41.42 influence each other, i.e. the current I6.38 used for determining the position in the X direction, for example. Influenced by the size of I7 or electrodes 43.44,
Since the degree changes depending on the incident position of the light beam in the Y direction, the position coordinate X can no longer be expressed by equation (1), and as a result, the position detection characteristic The problem was that the beam position was detected broadly at the center and narrowly at the periphery, exhibiting non-linear optical position detection characteristics.

In order to improve such non-linearity problems in two-dimensional optical position detectors, the magazine "Optics" Vol. 12 No. 5 (19
1983), pages 367-373, a double-sided split type PSD with current splitting layers on the front and back sides and a Ge
A so-called vinction type PSD, in which a dividing resistor surface is surrounded by a wire resistor having a specific resistance value, has been devised and is already on the market, as proposed by Ar.

However, the former method requires two highly uniform divided resistance layers with a large leakage current of the detector, and therefore has the disadvantage of a low manufacturing yield.

In addition, the latter method is difficult to manufacture because it is necessary to strictly match the ratio between the sheet resistance and the line resistance of the divided resistance layer, and there is a problem that the optical position detection characteristics deteriorate if the matching is poor.

The present invention aims to solve the above problems and provide a two-dimensional PSD with good position detection characteristics.

[Means to solve the problem] The present invention relates to a first conductivity type substrate made of a crystalline semiconductor.

a second conductivity type layer formed in the central region of the first surface of the substrate;
two pairs of opposing strip-shaped electrodes formed along the sides on the peripheral edge of the second conductivity type layer or the first surface of the first conductivity type substrate adjacent to the peripheral edge; In a two-dimensional optical position detection device whose basic structure is an electrode formed on a second surface, a field effect transistor is interposed between a second conductivity type layer region functioning as a photocurrent splitting layer and the strip electrode, The above problems are solved by making it possible to electrically connect or disconnect the two, thereby eliminating the influence between the opposing electrodes.

[Operation] As described above, the present invention significantly improves the characteristics of the surface-divided two-dimensional optical position detector, which conventionally had large non-linearity in position detection.

Compared to conventional elements, the added structure is simple, so there are few additional manufacturing processes, and the added functions are not analog but digital on/off functions.
The design parameters and process parameters have a wide allowable range, and manufacturing is easy.

[Example] Fig. 1 is a diagram showing the structure of a PSD which is an embodiment of the present invention, in which (a) is a plan view, and (b) is a plan view of the PSD.
It is an AA sectional view of.

In FIG. 1, 1 is an n-type Si substrate as a first conductivity type substrate.
A single crystal substrate, 2 a P-type layer as a second conductivity type layer, 13 a metal electrode for current extraction, 4 a metal electrode 13 for current extraction
low resistance 10 to obtain good electrical contact with the p-type layer 2
5 is a low-resistance n+ layer for the backside electrode, 6 is a field oxide film, and 18.degree. 19 is a gate electrode consisting of a metal electrode for voltage application.

7 is an insulating film. A field effect transistor (hereinafter referred to as FET) is formed between the gate electrode 19, the p-type layer 2, and the region indicated by 10 of the insulating film 7, and the current dividing layer region indicated by 8 of the p-type layer 2 and the current dividing layer region indicated by 9. It functions as a switch that can electrically connect or disconnect the current extraction electrode area. That is, in the case of the structure shown in the figure, the transistor exhibits depletion type FET characteristics, and by applying a positive voltage to the gate electrode 19, the region of the p-type layer 2 directly under the gate electrode 19 becomes highly resistive, and becomes 8 and 8. Area 9 is electrically cut off.

That is, in FIG. 1, a positive voltage is applied to the back electrode 5 with reference to the current extraction electrodes 12, 13, 14, 15, which are strip-shaped electrodes kept at the same potential, and the device is operated as a PSD. In this case, the gate electrode 1 is a strip-shaped electrode.
By applying a positive voltage to 8 and 19, electrodes 12 and 13 are electrically decoupled from current splitting layer 8 and no photogenerated current can flow into these electrodes, so that the optical position sensing device It is equivalent to a one-dimensional PSD that has only 14 and 15 as extraction electrodes and detects the position in the vertical direction of the figure, and there is no adverse effect from electrodes 12 and 13, and the position coordinates can be calculated accurately using equation (1). shows. To find the positional coordinates of the light beam in the direction perpendicular to it, apply a positive voltage to the gate electrodes 16 and 17, which are strip-shaped electrodes, to electrically separate electrodes 14 and 15 from 8. The optical position detection device shown in FIG. or accurately calculated properties.

Therefore, while applying a positive voltage alternately to the pair of opposing gate electrodes to electrically disconnect the pair of current extraction electrodes,
Accurate two-dimensional position coordinates can be obtained by alternately obtaining position information from currents from electrically connected current extraction electrodes.

FIG. 4(a) shows the results of evaluating the characteristics of the two-dimensional PSD according to the present invention operated in this manner when a light beam is swung in a lattice pattern.

FIG. 4(b) shows the evaluation results of similar characteristics when the same device was operated in the same manner as the conventional device without applying a voltage to the gate electrode. As can be seen from FIG. 4(b), the nonlinearity of the position detection characteristic, which is wide at the center and narrow at the periphery, becomes significantly smaller in FIG.

It is also uniform in the peripheral area, and in the case of this device, the nonlinearity has been improved by more than two orders of magnitude.

The above description has been made of the Tebre and Y-John type FETs, as well as the type that uses the photocurrent splitting layer as it is as the source, drain, and channel, which has a simple structure and is easy to manufacture. Of course, it is also possible to employ a depletion type FET in which one or more of the channels is formed independently of the photocurrent splitting layer.

An example of this structure is shown in FIG. In this example, only the channel region 20 utilizes the photocurrent dividing layer as it is in the same way as the source region, and the drain is composed of a dedicated low resistance p+ region 2.

Furthermore, it is of course possible to use an enhancement type FET.

An example of this structure is shown in FIG. In this example, the source region 21 uses the photocurrent splitting layer as is, and the channel region 2
3 uses an n-type substrate l, and the drain is a dedicated low-resistance p
It consists of + territory area 2. In this case, the current dividing region and the current extraction electrode, which have no effective electrical connection when no voltage is applied to the gate electrode, are electrically connected by applying a negative voltage, and the direction of the voltage applied to the gate electrode is Although it is the opposite, functionally it is possible to perform the same operation as the depression type.

In addition, a polysilicon gate is used as the FET gate.
In addition, Schottky Cate type FET, Pn
Of course, it is also possible to use a junction type FET.

Further, although the above description has been made regarding the case where an n-type substrate is used, it is of course possible to use a p-type substrate. In this case, the above description refers to n-type.

All you have to do is replace the P type and reverse the voltage polarity if necessary.

Furthermore, the idea in this invention is not limited to Si as long as the material is crystalline, but also single-element semiconductors such as Ge, ■-V group compound semiconductors represented by GaAs and InP, and even these Semiconductors in which some of the elements are replaced with elements in the same group of the periodic table, and group II-VI compound semiconductors represented by CdTe and ZnS, and furthermore, in which some of these elements are replaced with elements in the same group in the periodic table. Of course, it is also possible to use other semiconductors.

An example of the manufacturing method will be described below with reference to FIG.

In FIG. 1, 1 is an n-type Si single crystal substrate. A 6,000-layer oxide film was formed by wet and dry thermal oxidation, and the oxide film in the area that will become the p-type layer 2 on the front surface and the oxide film on the back surface was removed by photoetching, and the 2,000-layer oxide film was dry-oxidized. Boron ions are implanted into the portion that will become the p-type layer 2 at a concentration of 2.5×10 12 /cm 2 .

Furthermore, the photoresist was patterned by photoetching, a window was opened in it, and boron was added at 3 ox l.
Ions are implanted to a concentration of O''/c■2 to form a low resistance p+ layer 4.
form. 3. Add phosphorus on the back side. Ox 1015/cm2
A low resistance n' layer 5 for the back side electrode is formed by ion implantation to a concentration of . The oxide film on the upper part of the low-resistance p' layer 4 is removed by etching, and the implanted impurity is activated by annealing at 1000°C. The oxide film thickness of 2,000 layers is reduced by 1,000 layers by etching. Aluminum is vapor deposited on the front surface, a current extraction electrode 13 and a voltage application gate electrode 19 are formed by photoetching, and the aluminum is sintered by heat treatment.

Finally, a large number of PSDs were created on one wafer.
The chip is divided into individual chips, mounted in a case with a transparent glass window, and completed with wiring.

Next, the operation of the device of the present invention will be explained.

In FIG. 1(b), the gate electrode 19 and the p-type layer 2
A region indicated by 10 of the insulating film 7 forms an FET, and functions as a switch that can electrically connect or disconnect the current dividing region indicated by 8 and the current extraction electrode region indicated by 9 of the P-type layer 2. . That is, for the structure shown in the figure, FE
T indicates depletion type FET characteristics, and gate electrode 1
By applying a positive voltage to 9, the movable charges in the region of the p-type layer 2 directly under the gate electrode 19 are depleted, resulting in high resistance, and the regions 8 and 9 become electrically is blocked by.

FIG. 5 shows how the current flowing through the p-type layer 2 changes when a voltage is applied to the gate electrode 19. That is, when the voltage of the gate electrode 19 is Ov, the current of 97u, A is 12
．． OV is almost 0.

In this case, the effective resistance value of the p-type layer 2 is 10.3 in the former case.
In the latter case, it was measured to be 294 MΩ, and it can be seen that in the latter case, it is sufficiently electrically isolated between 8 and 9.

In FIG. 1, the current extraction electrode 12 kept at the same potential
, 13, 14.15 (hereinafter, voltages are measured using this voltage as a reference) When a positive voltage, for example 6■, is applied to the back electrode 5 and a light beam is irradiated, the following occurs. The photocurrent flows through the current extraction electrodes 12, 13, 14, and 15 divided according to the position of the light beam. At this time, when a positive voltage, for example 12V, is applied to the gate electrodes 18 and 19, the regions of the p-type current splitting layer directly under these gate electrodes become highly resistive, and the electrodes 12 and 13 are electrically disconnected from the current splitting layer 8. Since the optical position detection device is disconnected and no photogenerated current flows to these electrodes, the optical position detection device can be used as a current extraction electrode.
1 to detect the vertical position of a figure with only 14 and 15
It is equivalent to the dimensional PSD, has no adverse effects from the electrodes 12 and 13, and exhibits a characteristic in which the position coordinates can be accurately calculated using equation (1).

Next, gate electrodes 18 and 19 are returned to OV, and gate electrode 1
If a positive voltage of 12V is applied to electrodes 6 and 17, they are electrically disconnected from electrodes 14 and 15 or 8 by a similar mechanism, and the optical position detection device has only 12 and 13 as current extraction electrodes. It is equivalent to a one-dimensional PSD that detects the position in the direction, and exhibits a characteristic that the position coordinates can be accurately calculated using equation (1) without any adverse effects from the electrodes 14 and 15.

In this way, alternating voltages as shown in FIG. 7 are applied alternately to the pair of opposing gate electrodes to electrically separate the pair of current extraction electrodes, while at the same time disconnecting them from the electrically connected current extraction electrodes. By alternately obtaining position information from the currents, accurate two-dimensional position coordinates can be obtained.

FIG. 4(a) shows a light beam emitted over an area of 4×4×2 of a square two-dimensional PSD with a distance of 4.8×2 between the current extraction electrodes according to the present invention operated in this manner. 50
The position detection characteristics are shown when the graph is swung in a grid pattern at intervals of gm. FIG. 4(b) shows similar characteristics when the same device is operated in the same manner as the conventional device without applying a voltage to the gate electrode.

As can be seen from Figure 4, the non-linearity of the position detection characteristic, which is wide in the center and narrow in the periphery as seen in (b), is caused by (
a) It becomes noticeably smaller, and the spacing of the checkered pattern or the central part 9
It can be seen that the surrounding areas are almost uniform. In the case of the element shown in the figure, when the center is the origin and the X axis is horizontal and the Y axis is vertical, the nonlinearity on both axes within the range of 4■■ is approximately 7% to 0.05%. has improved by more than two orders of magnitude.

The above explanation is for depression type FET.
Furthermore, we have described a type that uses the photocurrent splitting layer as it is as the source, drain, and channel, which has a simple structure and is easy to manufacture.
Of course, it is also possible to employ a depletion type FET in which two or more photocurrent splitting layers are formed independently.

An example of this structure is shown in FIG. In this example, the peripheral portion of the photocurrent splitting layer 2 is used as it is for the source and channel, and the drain is constructed from a dedicated low-resistance P layer 22. This structure was manufactured by changing the photoetching pattern of the oxide film for forming the low-resistance p layer 4 to the low-resistance 29 layer 22 shown in FIG. 2 in the manufacturing method described in Example 1.

Regarding the light beam position detection characteristics, almost the same results as those shown in FIG. 4(a) were obtained.

Furthermore, it is of course possible to use an enhancement type FET. An example of this structure is shown in FIG.

In this example, the peripheral part of the photocurrent splitting layer 2 is used as it is for the source region 21, the n-type substrate is used as it is for the channel region 23, and the drain is composed of a dedicated low-resistance p+ layer 22. In this case, when no voltage is applied to the gate electrode, there is no effective electrical connection between the current dividing region and the current extraction electrode, but the current dividing region and the current extraction electrode are electrically connected by applying a negative voltage, and are functionally different from the depression type. It is possible to perform the same operation. In the manufacturing method described in Example 2, this structure is made by changing the photo-etching pattern of the oxide film for forming the P-type layer 2 so that the peripheral part is placed in the source region 21 as shown in the p-type layer 2 in FIG. I made it by changing the fastening position. Regarding the light beam position detection characteristics, almost the same results as those shown in FIG. 4(a) were obtained.

[Effects of the Invention] As explained above, according to the present invention, the characteristics of the surface-divided two-dimensional optical position detector, which conventionally had large non-linearity in position detection, have been significantly improved.

In order to improve the problem of non-linearity in two-dimensional optical position detectors, we have developed a double-sided split type PSD with a current splitting layer on the front and back sides, which has been considered in the past, and a split resistor surface proposed by G.A.R. The so-called vinction type PSD is surrounded by a wire resistor with a specific resistance value, but the former has the drawbacks of large leakage current and low manufacturing yield, and the latter has the disadvantage of increasing the sheet resistance and wire resistance of the divided resistor layer. It is necessary to match the ratio exactly, and if the matching is poor, the optical position detection characteristics deteriorate, whereas the PSD according to the present invention has a simple structure added to the conventional element. There are only a few additions to the manufacturing process, and the added functions are not analog, but digital on/off capabilities, so the tolerance range for design parameters and process parameters is wide, and the design is simple and easy to manufacture. , can be manufactured with high yield.

Furthermore, as a side effect, by applying a DC voltage of sufficient magnitude to any pair of gate electrodes, it is possible to operate as a one-dimensional PSD in a desired direction.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the structure of a PSD that is an embodiment of the present invention, in which (a) is a plan view, and (b) is a plan view of the same.
2 is a sectional view of an example of a structure employing a depression type FET, which is another embodiment of the present invention,
FIG. 3 is a sectional view of an example of a structure using an enhancement type FET, which is still another embodiment of the present invention, and FIG.
) is a diagram showing the results of evaluation of the characteristics of the two-dimensional optical position detection device according to the present invention, and (b) of the same diagram shows the same operation as the conventional device without applying voltage to the gate electrode. Figure 5 shows how the current flowing through the current dividing layer changes when voltage is applied to the gate. Figure 6 (a) explains the conventional PSD. A plan view of the same figure (b) is a BB% sectional view of the same figure (a),
FIG. 7(c) is a diagram showing details of FIG. 7(a), and FIG. 7 is a diagram showing necessary gate voltage waveforms in the present invention. In the figure, 1: n-type Si single crystal substrate 2: P-type layer 4: low resistance 21 layer 5: low resistance n+ layer for back side electrode 6: felt oxide film 7: insulating film 8: current dividing layer region 9. Current extraction electrode region lO0 field effect transistor region 12.13, 14, 15: current extraction electrode 16.17,
18, 19: Gate electrode 20: P region for channel 21: Source region 22 Low resistance p0 region for drain 23: Channel region 32: P type layer 33: P type layer Light receiving surface 34.35: P'' layer 36: Insulating layer 37. 38: Metal electrode 39: N+ layer 40: Electrode 41. 42: 2 sides without electrode 43. 44: Metal electrode Representative Patent attorney 1) Haru Kitatake Figure (b) Figure 1 (a) Figure 4 (a) (b) Figure (b) Figure 4

Claims (3)

[Claims] (1) A rectangular crystalline semiconductor first conductivity type substrate, a second conductivity type layer formed in the central region of the first surface of the first conductivity type substrate, and a peripheral edge or periphery of the second conductivity type layer. Two pairs of opposing band-shaped electrode groups formed along each side on the first surface of the first conductivity type substrate adjacent to the end, and electrodes formed on the second surface of the first conductivity type substrate. A two-dimensional optical position detection device having a basic structure, in which a field effect transistor is interposed between the second conductivity type layer region functioning as a photocurrent splitting layer and the two pairs of strip electrodes, and both are electrically connected. What is claimed is: 1. A semiconductor device for optical position detection, characterized in that it can be electrically connected or interrupted. (2) The field effect transistor interposed between the second conductivity type layer region functioning as a photocurrent splitting layer and the strip-shaped electrode has a width of the entire width of the second conductivity type layer region or 80% or more of the width of the second conductivity type layer region. Consists of a second conductivity type source, drain, channel, and gate formed on the channel via an insulating film, and when no gate voltage is applied to the gate electrode, the resistance between the source and drain is effectively in a conductive state. 2. The semiconductor device for optical position detection according to claim 1, wherein the semiconductor device is a so-called depletion type field effect transistor in which the source and drain are effectively cut off by applying a necessary voltage to the gate. (3) When the field effect transistor interposed between the second conductivity type layer region that functions as a photocurrent splitting layer and the strip electrode has no gate voltage applied to the gate electrode, the effective resistance between the source and drain increases. It is a so-called depletion field effect transistor, which is in a conductive state when 2. The semiconductor device for optical position detection according to claim 1, wherein the field effect transistor is constituted by a band-shaped gate electrode. (4) The field effect transistor interposed between the second conductivity type layer region functioning as a photocurrent splitting layer and the strip electrode has a width of the entire width of the second conductivity type layer region or 80% or more of the width of the second conductivity type layer region. Consists of a second conductivity type source and drain, a first conductivity type channel, and a gate formed on the channel via an insulating film, and when no gate voltage is applied to the gate electrode, the effective resistance between the source and drain is The light source according to claim 1, wherein the transistor is a so-called enhancement field effect transistor, which is in a cut-off state and is effectively electrically connected between the resource and the drain by applying a necessary voltage to the gate. Semiconductor device for position detection.